Jason Westin, MD: Other variables that can predict for outcomes that are not good are high LDH [lactate dehydrogenase] or high inflammatory markers prior to doing CAR [chimeric antigen receptor] T-cell infusion, and that may get back to the same thing about bulk, about rapidly proliferative tumors. It may be that we’re all talking about the same variable here—that bad lymphomas behave badly. But that is another factor.

David Maloney, MD, PhD: We’ve been talking mostly about axicabtagene ciloleucel. Clearly, there is a large body of evidence that’s emerging. It’s very encouraging with long-term follow-up. That’s not the only product approved. Tisagenlecleucel is also approved in this exact same setting—basically after 2 prior therapies. Jason, do you want to tell us a little about the long-term data, or the real-world data that are emerging with that?

Jason Westin, MD: That’s based on the JULIET study. As you said, it was based on very similar criteria to the ZUMA-1 trial. It did not allow for primary mediastinal patients, which is 1 of the main differences in eligibility. As we alluded to before, did it have a longer manufacturing time frame? Bridging therapy was allowed on that study. As we’re talking about today, that may have had some impact on outcomes.

The early data from the JULIET study showed a complete response rate. It was a little lower than we’ve seen on the ZUMA-1 trial; however, it was still around 40%. What we’re seeing is that patients who achieved a complete response on JULIET are tending to keep that response. We have a poster at the ASH 2019 [American Society of Hematology Annual Meeting & Exposition] updating at around 30-some months follow-up, and what we’re seeing is a plateau for patients who had long-term response.

David Maloney, MD, PhD: What about the real-world data with this agent?

Jason Westin, MD: There are some data coming from this meeting, as well as from some of the registries. They are now reporting on some of the real-world data. It’s not quite as many patients as we’ve seen in the real-world data sets from our CAR T-cell consortium or from Dana-Farber [Cancer Institute] that was presented by Caron last year at the ASH meeting, but there are data that are being reported on at this meeting showing that results seem to be very similar to what was seen on the clinical trials. I agree with the comments earlier. It’s very encouraging to see real-world data where patients don’t necessarily meet any eligibility or all the eligibility criteria but were seen on the early studies, who are able to receive these therapies, able to survive the toxicities, and still have similar long-term outcomes. Obviously, more follow-up is needed on these real-world data sets, but so far so good.

David Maloney, MD, PhD: Now, 1 of the biggest issues we’ve seen in the adoption of tisagenlecleucel in the clinics has actually been difficulties with obtaining the drug in a timely fashion and within specifications. There was an interesting abstract about that today. Who would like to discuss that? Jason?

Jason Westin, MD: Well, not necessarily specific to that abstract, but in general, the viability threshold for the product on the clinical trial was 70%. With FDA approval, the viability required for releasing the product was 80%—a bit higher of a bar. Manufacturing challenges to date have been described outside this discussion today. It is well known that there have been some challenges in making the product. However, the most recent data I’ve heard are that they’re able to deliver a product to 90% of patients who were having apheresis. Some of that is through an access protocol for products that are not quite meeting viability but are able to receive it. At least 10% of patients are not able to receive the product, but it’s an ongoing challenge.

Caron Jacobson, MD: In regard to that specific abstract, what they were able to show was that very limited numbers of patients treated in the real world for whom they didn’t see any difference in outcomes for the patients had viabilities between 60% and 80%, and patients who were 80% and above. And so as a physician treating a patient, when you get an out-of-spec product, it gives you more confidence that you’re giving the patient an active product. And we should say that in Europe the approval is actually for 70% viability for release. But I think that 10% number of not being able to deliver a product is still pretty significant. With axicabtagene ciloleucel, it’s more like 1% to 2% in terms of not being able to deliver a product. Again, it gets to the fact that we could tease out differences among the different products and toxicity and efficacy all we want, but when you have a patient in front of you, it doesn’t matter if you can’t treat that patient, right? And so that’s going to be paramount until we have products that are equally available.

David Maloney, MD, PhD: But I will say that it’s not surprising about the viability issue, because the dose is actually calculated on viable cells. And so they’re getting viable cells. I think that’s a bit of an artificial bar that they are being forced to meet. Now, in clinical trials we’re pretty restrained regarding who we treat. People have to meet eligibility criteria. Clearly, there’s a perception of treating 80-year-old patients. Can they really tolerate axicabtagene ciloleucel? Is tisagenlecleucel maybe a little kinder, gentler? The issue is, it’s really hard to tease out, in the real-world setting, whether physicians are using these biases when they’re selecting products. What do you think you’ve seen in your real-world data?

Caron Jacobson, MD: We haven’t yet talked about some of the different kinetics of 4-1BB versus CD28 CARs, in terms of rapidity of expansion and that leading to possible differences in toxicity. But the bottom line is, the 4-1BB CARs seem to expand more slowly. That potentially leads to a more modified toxicity profile. And so you might suspect that you would take your older, frailer, and potentially sicker patients to a 4-1BB CAR, but you have to remember that if it is taking 1 or 2 more weeks to get that CAR, or 105 of them aren’t getting it, you’ve lost your window to be able to treat some of these patients. In my practice, for our sicker patients, when we are more worried about their survival through the manufacturing process, we’re actually taking them to axicabtagene ciloleucel. I think in my practice, that got flipped on its head.

Frederick Locke, MD: One of the interesting things we’ve seen in several of the large real-world data sets with axicabtagene ciloleucel is that the outcomes for older patients—defined as over age 65—are equivalent. In some cases, they are better than we see with the younger patients. And so, more analyses have to be done to see if the older patients just have better disease features or performance status that sort of accounts for that, but there may also be something in the biology of their T cells or something else that we need to better understand. So we don’t have an age limit for CAR T-cell therapy. We are starting to see physicians in our practice, in our group, make decisions on which CAR T-cell therapy to use for their lymphoma patients based on some of these factors that you’ve mentioned. Severe toxicity rates are lower with the tisagenlecleucel product. If you have a patient with a very poor performance status or a low ejection fraction or something else, some physicians are choosing to use that product. Of course, that’s weighed against the fact that, as we said, they’re less likely to get that product back and be able to give it to the patient.

The other thing I would add is, if we look at the complete response [CR] rates and durable response rates across the 2 FDA-approved products, and the third that was presented here at ASH, the CR [complete response] rates and durable response rates are very comparable across these products. Then it does factor in—about the toxicities and the ability to get these therapies when we’re making these choices.

Transcript Edited for Clarity